Submitted to: Small Ruminant Research
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 4/5/2004
Publication Date: 7/1/2004
Citation: Moore, J.E., Goetsch, A.L., Luo, J., Owens, F.N., Galyean, M.L., Johnson, Z.B., Sahlu, T., Ferrell, C.L. 2004. Prediction of fecal crude protein excretion of goats. Small Ruminant Research. 53:275-292.
Interpretive Summary: Estimates of metabolic fecal crude protein (MFCP) were remarkably consistent among equations from the entire database and subsets constructed by deletion of observations with high residuals. Similarly, estimates of true crude protein (CP) digestibility varied little among simple regression equations derived from data subsets with extreme outliers deleted, some of which were from browse-containing diets, and for multiple regression equations that accounted for browse-containing diets. A regression of digestible crude protein against dietary CP from a data subset with deletion of observations having residuals
Technical Abstract: A database of 622 treatment mean observations of the dietary concentration of CP and apparently digestible CP (DCP) from 146 publications was used to estimate true digestibility of CP and metabolic fecal CP (MFCP) in goats. A regression of DCP against CP with the entire database yielded the equation: DCP = 0.8566 x CP (%DM)-2.697 (r2 = 0.851, root mean square error = 1.58). There were some observations with lower than predicted DCP, some of which were with diets containing browse. Therefore, observations with residuals <1.58 were deleted, resulting in the equation: DCP = 0.8831 x CP (%DM) -2.67 (r2 = 0.952, root mean square error = 0.86; n = 562); estimates of MFCP and true CP digestibility were considered the Y intercept and slope, respectively. To address variables of the entire database with less than expected DCP, the database was split into a subset to develop equations (60% of observations), with inclusion of additional variables such as DM intake and dietary concentrations of forage and browse, and one to evaluate. However, multiple regression equations did not greatly improve prediction, with lower than predicted DCP appearing a consequence of depressed true CP digestibility rather than increased MFCP. In conclusion, for goats consuming diets without browse, 0.88 and 2.67% DM appear appropriate estimates of true CP digestibility and MFCP, respectively, similar to values for other ruminant species.